A tale of two fish: Comparative transcriptomics of resistant and susceptible steelhead following exposure to Ceratonova shasta highlights differences in parasite recognition

Abstract

Diseases caused by myxozoan parasites represent a significant threat to the health of salmonids in both the wild and aquaculture setting, and there are no effective therapeutants for their control. The myxozoan Ceratonova shasta is an intestinal parasite of salmonids that causes severe enteronecrosis and mortality. Most fish populations appear genetically fixed as resistant or susceptible to the parasite, offering an attractive model system for studying the immune response to myxozoans. We hypothesized that early recognition of the parasite is a critical factor driving resistance and that susceptible fish would have a delayed immune response. RNA-seq was used to identify genes that were differentially expressed in the gills and intestine during the early stages of C. shasta infection in both resistant and susceptible steelhead (Oncorhynchus mykiss). This revealed a downregulation of genes involved in the IFN-γ signaling pathway in the gills of both phenotypes. Despite this, resistant fish quickly contained the infection and several immune genes, including two innate immune receptors were upregulated. Susceptible fish, on the other hand, failed to control parasite proliferation and had no discernible immune response to the parasite, including a near-complete lack of differential gene expression in the intestine. Further sequencing of intestinal samples from susceptible fish during the middle and late stages of infection showed a vigorous yet ineffective immune response driven by IFN-γ, and massive differential expression of genes involved in cell adhesion and the extracellular matrix, which coincided with the breakdown of the intestinal structure. Our results suggest that the parasite may be suppressing the host's immune system during the initial invasion, and that susceptible fish are unable to recognize the parasite invading the intestine or mount an effective immune response. These findings improve our understanding of myxozoan-host interactions while providing a set of putative resistance markers for future studies.

Fig 1. Experimental diagram of the exposure conditions and subsequent sampling of steelhead.

Susceptible steelhead (green) and resistant steelhead (orange) were exposed to Ceratonova shasta for 24 hours and then each phenotype was separated and placed into triplicate tanks. Resistant fish had been previously fin-clipped as a means of identification. dpe = days post exposure.

Fig 2. Histological sections of resistant and susceptible steelhead intestine after exposure to Ceratonova shasta.

Susceptible fish intestine at (A) 7 days post exposure (dpe), (B) 14 dpe showing chronic inflammation (asterisks) throughout the submucosa, and (C) 21 dpe with inflammation present in all tissue layers and sloughing of necrotic epithelia (arrow). Resistant fish intestine at (D) 7 dpe and (E) 21 dpe. Mature C. shasta myxospore (arrow) in the intestine of susceptible fish at 21 dpe (F). Bars = 100 μm.

Fig 3. Relative quantity of Ceratonova shasta DNA present in the gills (1 dpe) and intestine (7, 14, and 21 dpe) of infected steelhead (Oncorhynchus mykiss).

Each symbol represents the average quantitative cycle (Cq) of 100 ng of DNA extracted from the whole tissue (gills or intestine) of one fish that was assayed in triplicate by qPCR. Six fish of each phenotype were assayed at each timepoint. Fish that tested negative were assigned a nominal Cq value of 41. Dashed red lines indicate the average Cq values obtained from 1 and 1000 actinospore standards.

Fig 4. Venn Diagram showing the number of genes differentially expressed in response to Ceratonova shasta infection in the gills of resistant and susceptible steelhead at 1 day post exposure.

Arrows indicate upregulation vs downregulation.

Fig 5

GO enrichment among the genes downregulated in the gills of resistant (A) and susceptible (B) steelhead at 1 day post exposure to Ceratonova shasta. Enriched gene ontology (GO) terms were grouped into functionally related nodes using the Cytoscope plugin ClueGO. Nodes are colored and grouped according to a related function and labelled by the most significant term of the group. Node size corresponds to the FDR-adjusted p-value of each GO term and is specific to each graph.

Fig 6. Differential expression results for susceptible fish at 14- and 21-days post exposure (dpe) to Ceratonova shasta.

A) Venn diagram indicating the number of differentially expressed genes overlapping at 14- and 21 dpe. Arrows indicate up- vs. downregulation. B) Volcano plot of differential gene expression for susceptible fish at 14 dpe. Each dot represents the average value of one gene across three biological replicates. Red indicates the gene was significant at the FDR-adjusted p-value and Log₂-Foldchange threshold, blue is significantly only by p-value, green only by Log₂-Foldchange, and gray were not significant by either metric. B) Same as (A), but for susceptible fish at 21 dpe.

Fig 7. Functional enrichment of biological processes among the genes differentially expressed in the intestine of susceptible fish at 14 days post exposure to Ceratonova shasta.

Enriched gene ontology (GO) terms were grouped into functionally related nodes using the Cytoscope plugin ClueGO. Nodes are colored and grouped according to a related function and labelled by the most significant term of the group. Node size corresponds to the FDR-adjusted p-value of each GO term and is specific to each graph. The analysis was conducted separately on upregulated (A) and downregulated (B) genes.

Fig 8. Functional enrichment of biological processes among the genes differentially expressed in the intestine of susceptible fish at 21 days post exposure to Ceratonova shasta.

Enriched gene ontology (GO) terms were grouped into functionally related nodes using the Cytoscope plugin ClueGO. Nodes are colored and grouped according to a related function and labelled by the most significant term of the group. Node size corresponds to the FDR-adjusted p-value of each GO term and is specific to each graph. The analysis was conducted separately on upregulated (A) and downregulated (B) genes.

Fig 9. qPCR validation of RNA-seq results.

Quantitative reverse transcription PCR (RT-qPCR) validation of four immune genes (IFN-γ, TNF-α, IL-10, IL-1β) found to be significantly differentially expression by RNA-seq at day 7 in the gills. The X-axis shows the gene and phenotype assayed and the Y-axis shows the relative log₂(Fold Change) between fish exposed to Ceratonova shasta and their respective control. Error bars indicate the standard deviation of Cq values between biological replicates.